Hotel Lobby AI Prompt

A hotel lobby AI prompt is the instruction you give an agent when you are not using a one-click template. The composer on this site already injects a locked prompt; you only upload two photos. If you are driving MiniMax H3 / Kie / another agent yourself, paste the block below.

Note: this page covers the video trend only. The name is the song and the stage. If you want software to design or render hotel interiors, this isn't the page for you.

Prefer the preset instead? That is the hotel lobby AI template. Hub: hotel lobby AI video. Generate here: hotel lobby AI video generator.

What the prompt must lock

• Two subjects. @Image1 / Image 1 = left. @Image2 / Image 2 = right. Do not swap sides.
• Reference video is the master: orange COLORS studio, hanging mic, choreography, original audio.
• No new vocals, no marble lobby, no extra people, no captions.

Prompt you can paste

Short version (when the agent already has the reference video attached):

Replace BOTH COLORS “Hotel Lobby” performers. @Image1 = LEFT. @Image2 = RIGHT. Keep orange studio, hanging mic, choreography, original audio. No new vocals, no marble lobby, no side swap.

Full version:

STRICT TWO-SUBJECT REPLACEMENT.

Use the reference video as the master. Replace BOTH original performers completely:

@Image1 = LEFT performer.
@Image2 = RIGHT performer.

The references may be people, animals, or characters. Preserve each subject’s recognizable face/head, body shape, hair or fur, clothing, colors, and accessories. Keep both identities stable and never swap sides.

Preserve the reference video exactly: orange COLORS studio, hanging microphone, camera, framing, choreography, gestures, expressions, positions, timing, lighting, shadows, occlusion, and motion. Follow the video’s original medium or waist-up framing; full-body reference images are identity references only.

Retarget the original mouth motion and lip-sync timing to both replacements. Preserve the original audio, voices, music, beat, and sound effects unchanged. Do not generate new speech, vocals, or music.

Replace the complete visible appearance of both performers—not face overlays. Neither original performer may remain recognizable.

No identity drift, morphing, side switching, new movement, extra subjects, scene changes, text, subtitles, logos, or watermarks.

Final result: the same original Hotel Lobby COLORS performance, with only the two performers replaced by @Image1 and @Image2.

Attach the reference performance as video and the two photos as images. Duration should match the reference (5–15 seconds on this product).

Template vs prompt

Walking-through-a-hotel prompts will not produce this trend. If the vendor rejects the output as sensitive, change photos or retry — the same text can pass on one run and fail on another.